Audio-Driven Facial Animation Technology for NVIDIA Audio2Face

, Senior Research Scientist , NVIDIA
, Sr. Manager, Research, NVIDIA
We'll describe the audio-driven facial technology that's used for NVIDIA Audio2Face. From its beginning to current models, we'll explain how it evolved and which techniques contributed to get to the current high-quality model. We also explain how the technique is deployed as product. Finally, we introduce the future direction of Audio2Face technology.
活动: GTC 25
日期: March 2025
行业: 媒体与娱乐
话题: Simulation / Modeling / Design - Avatar / Digital Human
级别: Technical – Beginner
NVIDIA 技术: TensorRT,Audio2Face,Avatar Cloud Engine (ACE),cuBLAS,cuDNN
语言: 英语
所在地: